Motor learning and crossmodal interactions within the human mirror-neuron system in fMRI

The mirror-neuron system is assumed to be the neural basis for a direct
matching between observation and motor representations of actions and is
considered to mediate the understanding of actions and intentions (see [1]).
There seems to be a similar system for the auditory domain that matches a heard
action to the according motor representation (see e.g., [2]). However, the
crossmodal interactions within the system have not yet been investigated
systematically.
[1] Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual
Review of Neuroscience, 27, 169-192.
[2] Gazzola, V., Aziz-Zadeh, L., & Keysers, C. (2006). Empathy and the
somatotopic auditory mirror system in humans. Current Biology, 16, 1824-1829.

This project aims at investigating the crossmodal interactions and crossmodal
transfer during the perception of different newly learned motor patterns using
fMRI. Before scanning, participants will learn to perform key sequences on a
piano keyboard either within a purely auditory learning condition or within a
purely visual learning condition.
Firstly, it will be examined whether the auditory and the visual mirror system
share common neural substrates.
Secondly, we test which brain areas and which parts of the mirror system
respond to the sound of an action (i.e., a melody), if this action was learned
by observation of according finger movements. Correspondingly, it will be
tested, which brain areas and especially which parts of the mirror system are
activated during the observation of an action (finger movements on a piano
keyboard) which was learned to perform by listening to the corresponding
melody. In other words, we investigate whether there is a crossmodal transfer
if one action was learned within one modality. Thirdly, we examine whether
there are trimodal areas, which respond to the sound as well as to the
observation of an action. Furthermore, structural as well as functional data
will be related to behavioural data: Differences in white matter pathways in
each single participant will be examined using diffusion tensor imaging, in
order to correlate the obtained values to differential performance parameters
during the motor training and individual hemodynamic responses in relevant
brain areas during a (motor) perception task.

There are two different learning conditions:
1. observing finger movements showing a key sequence, while participants see
their own fingers during reproduction, but can*t hear a sound (learning by
vision only) or
2. listening to a key sequence (i.e., a melody), without seeing their own
finger movements during training, but hearing the result of own key presses
(learning by listening only).
After completing each of one training session, participants will take part in
scanning sessions including functional magnetic resonance imaging (fMRI) and
diffusion tensor imaging (DTI) after three days training and fMRI and an
anatomical MRI scan after 6 days training.
Within the fMRI scan they will hear (within one trial two, either same or
different) melodies, which they have learned to perform during the auditory or
the visual training or which they haven*t learned to perform. Accordingly,
participants will observe different finger movements which they have learned to
perform during the auditory or the visual training or which they haven*t
learned to perform.
This results in 6 different experimental conditions (for the first fMRI scan,
after visual or the auditory training only, there are 4 conditions,
respectively):
1. Vtrain Vtest: observing sequences that were previously trained to perform
visually
2. Atrain Vtest: observing sequences that were previously trained to perform
auditorily
3. no-train Vtest: observing sequences that were not trained to perform
4. Atrain Atest: hearing sequences that were previously trained auditorily
5. Vtrain Atest: hearing sequences that were previously trained visually
6. no-train Atest: hearing sequences that were not trained to perform
After that, participants receive a further training session during which they
are trained to perform all melodies presented auditorily and visually
simultaneously. During a third fMRI scan, within each trial, two (either same
or different) of the trained key sequences will be presented either both
visually, both auditorily or one visually and one auditorily.

Participants will be exposed to a 3 T magnetic field. No side effects have been
described so far. On rare occasions a peripheral nerve (abdomen) is stimulated
by the changing magnetic gradients. This will cause an itchy feeling, but is
not harmful.